The growing number of persons with heart failure (HF) in the United States is at high risk of two clinically catastrophic outcomes: unnecessary hospitalization and premature death. Those who are hospitalized increasingly are discharged to home health care where, despite a decade of national initiatives, the rate of rehospitalization remains stubbornly high. The days immediately after hospital discharge are a particularly vulnerable time for HF patients and important new research findings suggest that now is the time to take advantage of clinical practice variation to examine the effectiveness of early, intensive home health and physician services in reducing rehospitalization rates. The overall goal of the project is to improve HF patient outcomes by identifying the most effective combination of home health nursing visits and physician follow-up in reducing rehospitalization, and to determine whether some patients benefit more than others. The specific aims to accomplish this goal are: 7 Aim #1. To compare the effectiveness of higher versus lower intensity home health nursing services and physician follow-up during the week after hospital discharge on 30-day rehospitalization rates; and 7 Aim #2. To identify the extent to which HF severity, important clinical conditions (e.g., multimorbidity, cognitive functioning) and other patient factors modify the effect of early, intensive home health care and physician visits on 30-day rehospitalization rates. An analytic file of all hospitalizations for HF of Medicare fee-for-service beneficiaries discharged to home health care will be constructed by linking Medicare patient assessment, claims and other administrative data. Regression models of rehospitalization will be estimated to examine the influence of early, intensive home health and physician services while controlling for a wide range of patient, provider and market characteristics. Our primary outcome will be Ambulatory Care Sensitive Condition (ACSC) rehospitalizations because they are expected to be particularly sensitive to more effective home health and physician care. Secondarily, we will examine rehospitalizations specifically for HF as well as all-cause rehospitalizations. Instrumental variables estimation will be used to correct for potential endogeneity that arises from the non-random assignment of patients to different levels of home health care and physician follow-up, to produce unbiased estimates of the effect of these services on patient outcomes. The evidence emerging from this real-world comparative effectiveness study has the potential to rapidly change clinical practice and directly benefit medical decision makers, policymakers, and the growing number of older persons with HF who struggle to manage their disease and avoid catastrophic outcomes.